October 16, 2012
A new, larger transient elastography probe measured liver stiffness effectively in patients with nonalcoholic fatty liver disease, but its accuracy and reliability were reduced among obese patients in a recent study.
Researchers used M and new XL probes to perform liver stiffness measurement (LSM) on 193 patients with nonalcoholic fatty liver disease (NAFLD) in France and Hong Kong. The XL probe is larger, uses a lower ultrasound frequency than the M probe and is intended to improve measurement success among obese patients. Biopsy was performed within 24 hours after LSM, and liver histology served as reference standard to determine LSM accuracy.
Overall accuracy was similar between probes, and measurements correlated well (r2=0.9025, P<.001), but the XL probe often indicated lower LSM measurements (80% of patients). AUROC analysis produced the following values according to fibrosis staging:
- F2 or higher: 0.83 for M probe; 0.80 for XL
- F3 or higher: 0.87 for M; 0.85 for XL
- F4: 0.89 for M; 0.91 for XL
Investigators established an ideal cutoff of 7.2 kPa for F3 disease or higher, with a sensitivity and specificity of 78%, a PPV of 60% and NPV of 89%.
The XL probe obtained 10 valid acquisitions (95% of cases compared with 81%, P<.001)and a success rate greater than 60% (90% of cases compared with 74%, P<.001)more frequently than the M probe. Failure occurred in 10% of patients with the M probe vs. 2% with the XL (P=.002).
Success rate differences were more pronounced among obese patients, with 10 valid acquisitions obtained in 93% of patients compared with 60% using the M probe (P<.001).
Discordance of two or more stages between histology and XL probe results occurred in 9% of patients. Investigators observed an independent association between discordance and a BMI greater than 35 kg/m2 (aOR=9.09; 95% CI, 1.10-75.43).
“XL probe achieved successful and reliable LSM in the majority of NAFLD patients … but obesity is associated with less accurate and reliable measurements,” the researchers wrote. “With high negative predictive value, the technique can be used to exclude advanced fibrosis and cirrhosis reliably in routine clinical practice.”
Disclosure: See the study for a full list of relevant disclosures.